The Mid-Year Convention in a DCF: Why It Lifts Value by Roughly Half a Year of Discounting, and the Terminal-Value Trap That Catches Most Candidates
Michael King, PE Investment Manager · 9 min read ·
- The mid-year convention discounts each year's cash flow at period t-0.5 instead of t, on the logic that a business collects cash across the whole year, so the average pound arrives mid-year, not on 31 December
- The effect is a clean, uniform uplift of roughly (1+WACC)0.5 on the explicit cash flows — about 4.9% at a 10% WACC, rising with the discount rate
- The terminal value is where it goes wrong: a perpetuity-growth TV shifts to period N-0.5, but an exit-multiple TV stays at period N because it prices a year-end sale, not a stream of mid-year cash
- It is a convention, not a truth — its job is consistency with how comps and precedent multiples are typically struck, not a claim that cash literally arrives on 30 June
The Convention Corrects a Calendar Assumption, Not a Cash Flow
A standard DCF discounts year one's free cash flow by one full period, year two's by two, and so on: the exponent on the discount factor is the year number. That embeds an assumption almost nobody states out loud — that the entire year's cash arrives in a single lump on the last day of the year. It does not. A company bills, collects, and pays across all twelve months, so the cash-weighted average moment of collection sits near the middle of the year, not the end.
The mid-year convention fixes the exponent to match. Year one's cash is discounted at 0.5 periods, year two at 1.5, year three at 2.5, and so on — each flow pulled forward by half a year because, on average, that is when it lands. Nothing about the forecast changes. The cash flows are identical; only the timing assumption used to discount them moves.
The Uplift Is (1+WACC)0.5 — About 4.9% at a 10% Discount Rate
Because every explicit cash flow moves by the same half-period, the effect is not messy — it is a single multiplier applied to the whole explicit stream. Discounting at t-0.5 instead of t multiplies each present value by (1+WACC)0.5. At a 10% WACC that factor is 1.100.5 = 1.0488, an uplift of 4.9%. At 8% it is 3.9%; at 12%, 5.8%. The higher the discount rate, the more a half-year of earlier receipt is worth, so the more the convention adds.
A Worked Example: £100 of Flat Cash Flow, 10% WACC
Hold the cash flow flat at £100 per year for five years to isolate the timing effect, with a 10% WACC and 2% terminal growth. Under year-end discounting:
Now apply the mid-year convention. The explicit flows discount at 0.5 through 4.5, and — the point most candidates miss — the perpetuity-growth terminal value shifts back with them, to period 4.5:
The Terminal-Value Trap: Perpetuity Shifts to N-0.5, Exit Multiple Stays at N
This is where the convention earns its reputation as a favourite exam question, because the two terminal-value methods behave differently and the reason is not arbitrary.
A perpetuity-growth terminal value capitalises the company's ongoing cash flows into the future. Those cash flows are themselves subject to the mid-year assumption — each future year's cash arrives mid-year — so the perpetuity is valued half a year earlier than the year-end reference point. On a ten-year model, the Gordon-growth TV sits at period 9.5, not 10.0. It gets the same half-year pull-forward as everything else.
An exit-multiple terminal value does not. It prices a sale of the whole business at the end of the final forecast year — a single transaction on a single date, the last day of year N. A sale is a point-in-time event, not a stream of cash spread across a year, so there is no half-year of mid-period collection to reward. The exit-multiple TV is discounted at the full period N regardless of whether the convention is switched on.
| Component | Year-end discount period | Mid-year discount period | Why |
|---|---|---|---|
| Explicit FCF, year t | t | t − 0.5 | Cash collected across the year averages to mid-year |
| Perpetuity-growth TV | N | N − 0.5 | Capitalises a stream of mid-year cash flows, so it shifts with them |
| Exit-multiple TV | N | N | Prices a year-end sale — a point-in-time event, no half-year to reward |
Stub Periods Are the Same Idea, Applied to a Broken First Year
The convention assumes each forecast year is a full twelve months. Real valuation dates rarely land on a fiscal year-end, which leaves a partial first period — a stub. Value a company on 30 September with a December year-end and only three months, a 0.25 stub, remain in the current fiscal year. The first cash flow is scaled to that quarter, and under mid-year logic it is discounted at the midpoint of the stub — roughly 0.125 periods — with each subsequent full year offset from there. The principle is unchanged: discount cash at the middle of the window over which it is actually earned. The stub simply makes the first window shorter than a year.
It Is a Convention, So Use It Where the Comparison Demands It
Nothing about the mid-year convention is a law of finance. Cash does not literally arrive on 30 June. The convention is a modelling choice whose value is consistency: exit multiples are struck against trading and precedent comps, and if those reference multiples were themselves derived on a mid-year basis, a year-end DCF is comparing two differently-timed numbers. The convention exists to line the DCF up with the rest of the valuation toolkit, not because it is more true. The discipline that matters is not whether it is switched on, but whether it is applied consistently across the explicit period and the correct terminal-value method — and disclosed, so the reader knows a 5% chunk of the value is a timing assumption rather than a forecast.
Switch it on and forget the terminal-value distinction, and the model quietly double-counts or contradicts itself in a way that survives right up until someone senior checks the discount-period row. Switch it on knowingly, and it is a two-line adjustment that makes a DCF speak the same timing language as the comps sitting next to it on the football field.
Take Your Preparation Further
The mid-year convention only matters once the rest of the DCF is right, so read it against the pieces it plugs into. Start with DCF terminal value for the Gordon-growth-versus-exit-multiple choice this whole distinction turns on, and how to calculate WACC for the discount rate that sets the size of the uplift. For the cash-flow number being discounted, see unlevered free cash flow; for turning the enterprise value the DCF produces into a share price, the EV-to-equity bridge; and for the judgement that triangulates a DCF against the comps beside it, how to think about valuation.
For a model with the convention already wired into the discount-period row, download the DCF Model Template, and for the multiples that anchor an exit-multiple terminal value, the free Valuation Methods Cheat Sheet.
Ready for personalised feedback? Book a 1-on-1 mentoring session with an experienced IB/PE professional.
Frequently asked questions
What is the mid-year convention in a DCF?
It is a timing assumption that discounts each forecast year's free cash flow as if it arrives in the middle of the year rather than on the last day. Instead of discounting year one at period 1, year two at period 2 and so on, you discount at 0.5, 1.5, 2.5 and so on. The rationale is that a business collects cash across all twelve months, so the cash-weighted average moment of receipt sits near mid-year, not at the fiscal year-end. It changes only the discounting, never the underlying cash flows.
How much does the mid-year convention increase a DCF valuation?
On the explicit cash flows it is a uniform uplift of (1+WACC)^0.5, because every flow is pulled forward by the same half-period. At a 10% WACC that is 1.10^0.5 = 1.0488, or roughly 4.9%. The effect scales with the discount rate — about 3.9% at an 8% WACC and 5.8% at 12% — since a half-year of earlier receipt is worth more when money is discounted harder. Whether the whole enterprise value moves by that full percentage depends on how the terminal value is handled.
Does the terminal value change under the mid-year convention?
It depends on the method. A perpetuity-growth (Gordon) terminal value capitalises a stream of future cash flows that are themselves mid-year, so it shifts back half a period — on a ten-year model it sits at period 9.5 rather than 10.0 and picks up the same uplift as the explicit flows. An exit-multiple terminal value does not move: it prices a sale of the business at the end of the final year, a point-in-time event with no half-year of collection to reward, so it stays discounted at the full final period. Mixing these up — pulling an exit-multiple TV to N-0.5, or leaving a Gordon TV at N — is the single most common mid-year error.
Why does the exit-multiple terminal value stay at the full period?
Because it represents a discrete transaction, not a flow of cash. The exit multiple applies to the final forecast year's metric — say terminal EBITDA — and assumes the whole company is sold at the end of that year. A sale happens on a date, so its proceeds arrive at year-end, period N. There is no twelve-month collection window to average to the middle, which is exactly what justifies the half-year shift for the explicit cash flows and for a perpetuity. So the exit-multiple TV is discounted at N whether or not the mid-year convention is switched on.
Is the mid-year convention just a way to inflate a valuation?
It raises the number, but that is not its purpose and it is not a trick when used properly. The convention exists to make the DCF's timing consistent with how the rest of the valuation is built — trading comps, precedent transactions and the exit multiples drawn from them. Applied consistently across the explicit period and the correct terminal-value method, and disclosed so the reader knows a few percent of the value is a timing choice, it is standard practice. It becomes a problem only when it is switched on selectively, contradicts the terminal-value treatment, or is used to quietly manufacture a higher headline value the forecasts do not support.